Journal of the Neurological Sciences 350 (2015) 51–60
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Clinico-radiological spectrum and outcome in idiopathic hypertrophic pachymeningitis Gopal Krishna Dash, Bejoy Thomas, Muralidharan Nair, Ashalatha Radhakrishnan ⁎ Department of Neurology and Imaging & Interventional Radiology, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Trivandrum, India
a r t i c l e
i n f o
Article history: Received 31 October 2014 Received in revised form 25 January 2015 Accepted 5 February 2015 Available online 14 February 2015 Keywords: Hypertrophic pachymeningitis Clinical features Radiological features Treatment modalities Outcome
a b s t r a c t Objective: To elucidate the clinico-radiological features, treatment response and outcome of a large cohort of patients (n = 20) with idiopathic hypertrophic pachymeningitis (IHP) and to examine if any of these features could differentiate between IHP and secondary causes of hypertrophic pachymeningitis (SHP). Methods: 20 patients with IHP diagnosed between 1998 and 2009 formed the study cohort. We adopted a validated clinical score to quantitatively assess and document their neurological disability and to compare their pre- and post-treatment outcomes. Appropriate statistical analysis was done to look for any clinical and/or radiological features to differentiate IHP from SHP. Results: Out of the twenty eight consecutive patients with pachymeningitis, 20 were having IHP and 8 were having SHP (Tuberculosis-5, Sarcoidosis-2, Wegener's granulomatosis-1). In IHP, headache and visual symptoms dominated the clinical symptomatology (80% and 75%). In MRI, the peripheral pattern of contrast enhancement was more common with IHP (p = 0.03). The posterior falx and tentorium showing a hypointense center (“fibrosis”) and enhancing periphery (“active inflammation”) together mimicking “Eiffel-by-night” sign was found to be more commonly associated with IHP (60% vs 12.5%, p = 0.03). Biopsy was done in 9 patients. At a mean follow-up of 51 months (range 24–144 months), the mean pretreatment clinical score improved from 6.55 to 1.80 in 20 patients with IHP (p b 0.001). Conclusions: Our data on the largest cohort of patients with IHP would shed light into its clinico-radiological spectrum, treatment and outcome. The prognosis is satisfactory if managed appropriately. We have highlighted the role of MRI in differentiating between IHP and other causes of SHP. © 2015 Published by Elsevier B.V.
1. Introduction Hypertrophic pachymeningitis (HP) is characterized by localized or diffuse thickening of the cranial or spinal dura mater which may or may not be associated with inflammation, resulting in progressive neurological deficits [1,2]. It has been described in association with infection, trauma, tumors, and Wegener's granulomatosis [3–5] (Appendix 1). IgG4-related hypertrophic pachymeningitis (IgG4-RHP), a recently described entity, is an increasingly recognized manifestation of IgG4-related disease, a fibroinflammatory condition that can affect virtually any organ. It is estimated that IgG4-RHP may account for a high proportion of cases of hypertrophic pachymeningitis once considered idiopathic [6,7].
⁎ Corresponding author at: Department of Neurology, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Trivandrum, Kerala, India-695011. Tel.: + 91 471 2524684; fax: +91 471 2446433. E-mail address: [email protected] (A. Radhakrishnan).
http://dx.doi.org/10.1016/j.jns.2015.02.013 0022-510X/© 2015 Published by Elsevier B.V.
From the management point of view, attempts to differentiate primary or idiopathic HP (IHP) from secondary hypertrophic pachymeningitis (SHP) is very important. Definite diagnosis can be obtained after meningeal biopsy, but may not always be possible or warranted. MRI characterizes the degree of dural inflammation and clinches the diagnosis of HP, but till date there is no data on how the imaging helps in differentiating IHP and SHP. Most available literature on IHP being either case reports or small series, focusing on MRI abnormalities or the variable clinical manifestations and the clinical outcome with short follow-up, one cannot deduce any conclusive data on the same. No objective scales assessing the disability have ever been used to quantify the long-term functional outcome of a large group of IHP patients. This is especially important because of the prevailing notion that a majority of these patients do not recover fully and it is a long-standing, chronic disease with remission and relapses. Therefore, we attempted to elucidate the clinico-radiological and laboratory features, treatment response and outcome of a large cohort of patients with pachymeningitis. We specifically examined if any of these features could distinguish IHP from SHP thus aiding the treating neurologist.
52
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
2. Methods
2.4. Brain biopsy
2.1. Subjects and methodology
Biopsy was performed in 9 patients and included meninges and/or brain parenchyma. They were stained for infectious agents and cultures were done.
This study was conducted at the Sree Chitra Tirunal Institute for Medical Sciences and Technology, a tertiary center for neurological disorders in India. Institutional ethics committee approval was obtained for the study. 28 consecutive patients with hypertrophic pachymeningitis diagnosed between January 1998 and December 2009 were evaluated in detail. We adopted a simple composite clinical score derived from the various clinical manifestations in patients with hypertrophic pachymeningitis available in published literature and then compared the pre- and post-treatment score to evaluate their outcome homogeneously. Each component of the score has been adopted and modified from previously validated clinical scores [8–11]. The details of the clinical scores are provided in Appendix 2a. The authors (GKD and AR) did a detailed chart review to score each of the patient's status at various periods of follow-up, this being a simple scale and the scoring could easily be accomplished in all patients. Each patient was then personally interviewed by the Neurologists (GKD and AR) and patients were also made to score their symptoms before and after treatment in a Likert scale independently (Appendix 2b). The variability if any, in each of the score pertaining to a specific symptom and the composite score as scored by the authors and the patients were statistically analyzed by kappa statistics.
2.2. Investigations All patients underwent the following investigations: hemogram (hemoglobin, differential and total leukocyte counts, total platelet count), erythrocyte sedimentation rate (ESR), biochemical investigations including serum calcium, phosphorus, liver function tests, renal function tests, serology for human immunodeficiency virus (HIV), rheumatoid factor(RA factor), antinuclear antibodies, APLA (antiphospholipid antibody),serum VDRL (venereal disease reference laboratory testing),P and C antineutrophil cytoplasmic antibodies (cANCA and pANCA) and angiotensin converting enzyme(ACE) assays. Cerebrospinal fluid (CSF) examination was done in all patients for cytology, biochemistry and immunological tests to exclude secondary causes. A Mantoux test was performed in all patients.
2.3. Imaging Imaging studies included chest X-ray and contrast enhanced MRI brain (1.5 T Scanner, Signa GE, Milwaukee, WI) in all patients. All the MRI scans archived in our system were analyzed by the neuroradiologist (BT) blinded to the clinical data. Follow-up contrast enhanced MRI was performed if clinically indicated, and the findings were classified as (i) improved, (ii) worsened, or (iii) no change in relation to the immediately prior MRI(s). In the final follow-up MRI, the degree of abnormal enhancement was defined as (i) stable when unchanged, (ii) improved when there was less extensive enhancement or reduced thickness of the dura mater, and (iii) worse if there was increased enhancement compared to the initial imaging. [12] The patterns and sites of enhancement were noted and characterized as described by Hatano et al. in 1999 [13]. It was described as linear, nodular and combined. We also classified the pattern of dural enhancement as “peripheral”, “uniform” or “combined”. The involvement of pachymeninges was described as diffuse if more than 2 non-contiguous sites were involved and focal if less than 2 sites were involved and were also subclassified as “symmetric” or “asymmetric” by comparing either side of the brain.
2.5. Statistical analysis All analyses were performed using SPSS version 17.0 (SPSS Inc., Chicago, IL). Fisher's exact test, Paired t-test, kappa statistics, positive and negative predictive values, sensitivity and specificity were applied as appropriate. A p-value ≤0.05 was taken as significant. 3. Results 3.1. Clinical profile Out of 28 patients with HP, 20 patients had IHP and 8 patients had SHP. In the IHP group, there were 11 men and 9 women with a median
Table 1 Clinical features, CSF and MRI findings in patients with hypertrophic pachymeningitis (N = 28). Characteristics
IHP (N = 20)
SHP (N = 8)
Significance p*
Age, mean, years Sex (female:male) Age at onset of symptoms, mean, years Duration of illness, mean, years Clinical features Headache Visual symptoms Seizure Hearing loss Dysphagia Dysarthria Ataxia Facial sensory complaints Hemiparesis Quadriparesis Only second cranial nerve involvement Second cranial nerve involvement with other cranial nerve palsies Involvement of oculomotor nerves Facial nerve involvement Eight cranial nerve involvement Bulbar palsy CSF protein, mg% (mean, range) CSF glucose, mg% (mean, range) CSF cell count (mean, range) Site(s) of involvement in MRI Falcotentorial Medial frontal Basifrontal Frontal convexity Temporal convexity Skull base Symmetrical involvement in MRI Diffuse involvement Spinal involvement T2 Signal in MRI Hypointense Isointense Peripheral pattern of CE
49.5 9:11 44.5
40.8 5:3 39.7
NS NS NS
5.1 N, % 16(80.0) 15(75.0) 6(30.0) 4(20.0) 3(15.0) 2(10.0) 5(25.0) 2(10.0) 2(10.0) 2(10.0) 5(25.0)
1.28 N, % 7(87.5) 3(37.5) 2(25.0) 2(25.0) 1(12.5) 1(12.5) 2(25.0) 1(12.5) 0(0.0) 0(0.0) 2(25.0)
NS NS NS NS NS NS NS NS NS NS NS NS
5(25.0)
1(12.5)
NS
7(35.0) 2(10.0) 4(20.0) 5(25.0) 96.9(30–494) 82(50–154) 35(2–460) I 17(85.0) 7(35.0) 7(35.0) 6(30.0) 6(30.0) 5(25.0) 10(50.0) 13(65.0) 2(10)
3(37.5) 0(0.0) 2(25.0) 3(37.5) 76(37–159) 82.8(53–153) 16(2–75)
NS NS NS NS NS NS NS
4(50.0) 6(75.0) 4(50.0) 5(62.5) 3(37.5) 2(25.0) 4(50.0) 7(87.5) 0(0.0)
NS NS NS NS NS NS NS NS NS
19(95.0) 1(5.0) 12(60.0) 12(60.0) 3(15.0)
6(75.0) 2(25.0) 1(12.5) 1(12.5) 5(62.5)
NS NS NS 0.03 0.02
Follow-up MRI showing improvement
IHP — idiopathic hypertrophic pachymeningitis, SHP — secondary hypertrophic pachymeningitis, N — number, NS — not significant, * — p-value by Fisher's exact test, ^ — see text for description, CE — contrast enhancement.
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
53
Fig. 1. Graphical representation of clinical features in patients with idiopathic hypertrophic pachymeningitis in percentages.
age at presentation of 48 years (range 50–80 years). The details of their clinical features are provided in Table 1 and Fig. 1. The most common presentation in patients with IHP was headache (16 / 20, 80%). Poor visual acuity was noted in 9 patients with IHP, which ranged from 6/9 (Snellen's chart) to perception of light only. The most common cranial nerve involved in isolation was the second cranial nerve, seen in 5(25%) patients. No clinical feature could differentiate IHP from SHP (Table 1). At a mean follow-up of 51 months (range 24–144 months), the mean clinical score improved from 6.55(pre-treatment) to 1.80(post-treatment) in patients with IHP (p b 0.001), Fig. 2. The variability in clinical scoring between the patient and clinician remained “almost perfect” (Kappa varied between 0.8 and 1), except that in the severity of headache, the score showed only a “moderate agreement” (Kappa = 0.529) (Appendix 3). 3.2. CSF and other laboratory findings The CSF findings are summarized in Table 1. CSF parameters could not differentiate between IHP and SHP. In the IHP group, the mean ESR was 49 mm/h (range 4–124). RA factor and APLA were
positive in one patient each. CT scan of thorax and abdomen was normal in all patients with a diagnosis of IHP. In 4 patients with SHP, Mantoux test was positive and one patient had cANCA positivity.
3.3. Neuroimaging findings The details of the MRI findings in patients with HP are summarized in Table 1. In the IHP group, the most common pattern of pachymeningeal involvement was falcotentorial, seen in 17(85%) patients and the commonest pattern of contrast enhancement was a peripheral pattern (12, 55%) followed by a uniform pattern. Combined pattern (both peripheral and uniform) of enhancement was noted in only one patient, while nodular pattern was not seen. Associated spinal involvement was seen in 2 patients with IHP and none in those with SHP. Associated brain parenchymal involvement was noted in 6 patients (Fig. 3), venous compression in 6, leptomeningeal involvement in 3 and venous thrombosis in one patient. Significantly more number of patients with IHP had peripheral pattern of contrast enhancement (12/20 vs 1/8, p = 0.03) and they exhibited “Eiffel-by-night” sign [14] in MRI (12 vs 1, p = 0.03). This sign was described by our group earlier and it represents thickened dura in the posterior falx and tentorium due to fibrosis seen as central hypointensity in post-contrast fat saturated T1W image with an enhancing periphery depicting areas of active inflammation, mimicking the illuminated Eiffel tower at night (Fig. 4). The positive and negative predictive values, sensitivity and specificity of this sign in predicting the diagnosis of IHP were 92%, 46%, 60% and 87% respectively. Follow-up scans were available in 13 patients with IHP, the findings remained unchanged in 11 and showed progression in 2 patients during follow-up. More number of patients with SHP had improvement in their MRI scan as opposed to IHP group (3 vs 5, p = 0.02) (Fig. 5).
3.4. Histopathological findings
Fig. 2. Box plot depicting the total clinical scores of 20 patients with idiopathic hypertrophic pachymeningitis before and after treatment. Solid horizontal lines are median values, and length boxes are interquartile ranges (25th to 75th percentiles) excluding the outliers. The tip of the vertical lines from the top and bottom of the boxes indicates the maximum and minimum values excluding the extreme values (*).
Histopathological data was available in 9 IHP patients who underwent biopsy. Meningeal biopsies revealed dense fibrous infiltration with few plasma cells and lymphocytes. There was no granuloma formation, vasculitis or neoplastic cells (Fig. 6, patient 18). Stains for fungus, acid fast bacilli and cultures were negative in all cases of IHP.
54
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
Fig. 3. A) Axial fluid attenuated inversion recovery sequence image (FLAIR) of patient 8 showing significant bilateral occipital hyperintensity suggestive of edema. B) T1 weighted axial post contrast fat sat image showing contrast enhancement in the posterior falx (arrow). C) Axial FLAIR image of the same patient after one year of treatment showing significant resolution of edema and D) decrease in contrast enhancement.
3.5. Treatment and outcome The most common drug used for treatment in IHP was prednisolone. Other immunomodulatory agents used were azathioprine in 3, methotrexate in one, cyclophosphamide in one and mycophenolate mofetil in one patient respectively. During a mean follow-up of 51 months (range 24–144 months), headache recovered in 13(65%) patients and improved but not recovered fully in another 7(35%) patients. Except in 3(15%) patients, cranial nerve palsies recovered in all. There was a significant difference between the pretreatment and post-treatment score (6.55 vs 1.80, p b 0.001) (Fig. 2). In patients with SHP, 5 were treated with antitubercular drugs, one with prednisolone and two received combined prednisolone
and azathioprine(for sarcoidosis and Wegener's granulomatosis respectively). We have summarized the data of all 20 cases of IHP in Table 2. We have also compared the various large series of patients with IHP reported so far with ours in Table 3. 4. Discussion In this largest cohort of 20 patients with IHP, many patients were in their late fifties during presentation similar to previously reported series [12,15]. Both the genders were affected equally [2, 13]. In patients with IHP, headache was the most common presentation. Other small series are also in agreement with this observation [2,12,16]. Headache is due mainly to inflammation of the dura
Fig. 4. (A) T2-W coronal image at the falco-tentorial junction of patient 10 demonstrates the thick hypointense dura. (B) Contrast enhanced T1-W fat saturated coronal image through the same region shows thickened dura with hypointense center and enhancing periphery. (C) Contrast enhanced image 3 years later in the same patient shows persistence of contrast enhancement. (D) The peripheral enhancement pattern of the thick hypointense dura mimics the image of illuminated Eiffel tower (‘Eiffel-by-night’ sign).
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
55
Fig. 5. (A) Contrast enhanced fat saturated T1-W coronal image through the posterior falx and tentorium shows uniform enhancement in the tentorium, in a patient with secondary hypertrophic pachymeningitis due to tuberculosis (‘Eiffel-by-night’ sign negative). (B) Contrast enhanced image two years later in the same patient shows significant improvement of the dural enhancement.
mater, but can also be due to raised intracranial pressure [12,15]. The next most common manifestation was visual loss due to optic neuropathy, subsequent optic atrophy and the oculomotor nerve (i.e. third, fourth, and sixth cranial nerves) involvement. This is in contrast to a previously reported series comprising 12 patients who were the largest cohort described so far, where visual loss was seen in 11 out of 12 patients [12]. Such discrepancies are expected in rare diseases since no single center can have a large group of patients to audit the manifestations accurately. Apart from the second and oculomotor nerve involvement, bulbar palsy was noted in 25%, seventh and eighth cranial nerve involvement in 10 and 20% respectively. Some studies on IHP have described seventh nerve followed by lower cranial nerves as the most common pattern of cranial neuropathy [12]. However, we found that the involvement of optic and oculomotor nerves was more common. Seizures occurred in 30% and encephalopathy and hemiparesis in 20% each. Seizures albeit rare, can occur in patients with IHP related to parenchymal edema adjacent to the inflamed meninges or venous infarct. None of the clinical findings or routine laboratory features could
Fig. 6. Histopathology in IHP (patient 18) (hematoxylin and eosin stain ×500) showing the dura with dense fibrocollagenous tissue only.
differentiate IHP from SHP. This underscores the importance of a thorough evaluation by specific laboratory investigations, MRI and if needed biopsy in all patients with pachymeningitis since no clinical or laboratory parameter can sufficiently bolster ones clinical certainty as to the cause of HP. Contrast MRI of the brain and spinal cord is one of the key investigations in patients with HP which demonstrates the thickened and abnormally enhancing dura mater. It can assess the localization, extent, and pattern of involvement of abnormal dura and helps to monitor the progression. The sites of abnormal dural enhancement most commonly seen in patients with IHP were falco-tentorial (85%), medial and basifrontal (35%) followed by frontal and temporal convexities and skull base. We did not find a significant difference between the sites of involvement in IHP and SHP. A similar pattern of dural enhancement was noted in a previous series by Kupersmith et al [12]. However, in contrast to their series, we did not observe a dominant involvement of cavernous sinus. The authors have pointed out that the abnormal enhancement of sphenoid wing dura mater and cavernous sinus (100%) in their series might reflect a bias to patients with neuro-ophthalmologic symptoms since there was an ophthalmic center. We observed both symmetric and asymmetric pattern of dural enhancement occurring equally in our cohort (50% each), but more patients had a peripheral pattern of enhancement (60%). ‘Eiffel-by-night’ sign, described previously [14] by our group was noted in 60% of IHP patients, while only 12% with SHP had a similar MRI enhancement (p-0.03). In IHP, pathologically diffuse thickening of the dura with dense fibrosis and inflammatory cell infiltration more marked at the periphery composed of lymphocytes and plasmacytes was noted. This fibrosis imparts a hypointense signal on all sequences in MRI and the site of active inflammation on the surface is marked by contrast enhancement. This new sign encompassing all these features should therefore be specifically looked for in all cases of IHP and if found points towards IHP. We could not account for the pathogenesis of this specific pattern of involvement in IHP. Similarly, significant improvement in the degree of pachymeningeal thickening in MRI after treatment occurs only in secondary causes of HP, whereas in IHP, even with clinical improvement the same degree of MRI improvement in the form of reduction in the thickening of dura does not occur. At a median follow-up of 24 months, despite clinical improvement, 8 out of 20 patients continued to have similar imaging findings in IHP. Hence one should not be over enthusiastic in treating such cases with immunomodulators
56
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
Table 2 Summary of clinical, CSF, MRI findings, treatment and outcome of patients with IHP. Pt Age/Sex Clinical presentation No 1
51/F
2
53/M
3
40/M
4
64/M
5
57/M
6
43/F
7
50/M
8 9
54/F 30/M
10
35/M
11
39/F
12
67/M
13
38/F
14 15
49/F 37/F
16
47/M
17
61/M
18
38/F
19
80/M
20
58/F
Pre-treatment CSF clinical score* protein (mg/dl)
Headache and 3rd, 5th, and 6th CN 5 involvement Headache, seizures, and 2nd CN 5 involvement Headache, hemiparesis, and 2nd CN 7 involvement Headache, ataxia, and 2nd 8th CN 7 involvement Headache, seizure, and 2nd & 6th CN 7 involvement Headache, encephalopathy, and 2nd, 9 3rd, 6th, and 8th CN involvement Ataxia, hemiparesis 2nd, and 8th CN 5 involvement Headache and encephalopathy 2 Headache, seizure 2nd CN 7 involvement Headache, and 2nd, 3rd, & 6th CN 6 involvement Seizure, and 2nd & 3rd CN 5 involvement Headache, and 2nd, 3rd, 5th, 7th, 9th, 8 10th, and 11th CN involvement 6 Headache, ataxia, hemiparesis, and 2nd, 6th, 9th, 10th, 11th, and 12th CN involvement Headache, and 2nd CN involvement 10 Headache, seizures, and 6th CN 5 involvement Headache, and 2nd, 9th, 10th, 11th, 4 and 12th CN involvement Ataxia, hemiparesis, and 8th CN 6 involvement Headache, seizures, and 2nd CN 5 involvement Headache, and 2nd, 5th, and 6th CN 3 involvement Headache, and 2nd, 8th, 9th, 10th, and 11 11th CN involvement
CSF MRI* (Pattern of cells pachymeningeal thickening)
Eiffel-by-night Biopsy Treatment Follow-up sign* clinical score*
Follow-up duration, months
87
2
Diffuse, symmetric
+
No
Pred
1
24
32
2
Diffuse, symmetric
+
No
Pred
0
48
56
6
Focal, asymmetric
−
Yes
Pred
0
48
359
460
Focal, asymmetric
−
No
Pred
2
48
109
50
Diffuse, symmetric
+
Yes
Pred
0
18
42
2
Focal, asymmetric
−
Yes
Pred
5
144
84
40
Diffuse, symmetric
+
No
Pred
1
18
26 55
2 2
Diffuse, symmetric Focal, asymmetric
+ +
No No
Pred, Myc Pred
0 0
24 26
87
5
Diffuse, symmetric
+
Yes
Pred
3
24
59
3
Diffuse, asymmetric
+
No
Pred
0
40
54
5
Focal, asymmetric
−
No
Pred, Az
1
24
56
50
Focal, asymmetric
−
Yes
Pred
2
84
53 45
10 2
Focal, asymmetric Diffuse, symmetric
− +
Yes No
Pred Pred, Mtx
3 1
25 84
44
10
Diffuse, asymmetric
+
Yes
Pred, Az
1
84
494
20
Focal, asymmetric
−
No
Pred, Az
2
18
71
5
Diffuse, symmetric
+
Yes
Pred
2
18
55
10
Diffuse, symmetric
+
No
Pred
1
18
30
2
Diffuse, symmetric
+
Yes
Pred
0
14
IHP — idiopathic hypertrophic pachymeningitis, M — male, F — female, CN — cranial nerve, Pred — prednisolone, Az — azathioprine, Myc — mycophenolate mofetil, Mtx — methotrexate, * — see text for definition.
repeatedly even if the MRI is not showing an improvement pari passu with the clinical. Although biopsy is considered the gold standard for the definitive diagnosis of HP, the indications for biopsy is still not clearly laid down due to scarcity of literature. One can plausibly suggest that biopsy of the dural lesion should be considered if the patient diagnosed as IHP clinically deteriorates or the neuroimaging worsens despite treatment when the investigations fail to detect any secondary cause for HP. Chronic inflammation of the usually inert dura mater with minimal or no inflammation of the pia and arachnoid is seen in IHP. Lymphocytes, plasma cells, macrophages, epithelioid cells, and occasionally granulomas infiltrate the dura, later to be replaced by fibrosis. In the series by Kupersmith et al. on 12, two developed malignancy (one patient had carcinoma breast and the other had prostate malignancy) within a month of diagnosis of IHP. But such changes if destined to happen would become evident within few months and therefore a stringent follow-up of a minimum of one year is warranted. IgG4-related disease, a recently described condition with inflammatory meningeal disease accounts for an increasing proportion of cases of idiopathic hypertrophic pachymeningitis. Clinicians should become familiar with this alternative differential diagnosis and should appropriately test for the CSF and serum for the presence of immunoglobulin [6,7]. In addition to the MRI findings, CSF analysis, particularly CSF IgG (specifically
IgG4 levels) index and serum IgG analysis will unfold the etiological diagnosis of patients with inflammatory meningeal diseases [17]. These tests should be included while evaluating patients with hypertrophic pachymeningitis. Patients with IHP frequently progresses, if not managed properly. Steroid was the preferred drug in all our patients. Azathioprine, methotrexate and mycophenolate were used in few patients additionally (5 patients). During a mean follow-up of 51 months (range 24–144 months), headache recovered in 13 patients (65%) and improved, but not recovered fully in another 7(35%) patients. Except in 3(15%), cranial nerve palsies recovered in all patients. In a previous series of IHP by Masson et al., all 7 patients had relapse of symptoms after withdrawing the steroids [2]. Out of 12 patients in Kupersmith series, 7 out of 8 patients with visual loss improved rapidly and maintained their vision on therapy, suggesting that visual loss could be secondary to inflammation rather than the compressive fibrosis around the optic nerve [12]. Similarly, the gait imbalance, hemiparesis, and epilepsy also improved with corticosteroid treatment. Previous studies have shown that azathioprine or methotrexate allows tapering of steroids [12,10]. Hamilton et al., described 3 patients with optic neuropathy or painful ophthalmoparesis or both where the symptoms in two cases were controlled only by high dose prednisolone, until azathioprine was
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
57
Table 3 Comparison of the salient features in three large studies of hypertrophic cranial pachymeningitis. Parameters
Kupersmith (2004)
Riku S (2003)
Present study
Number of patients Mean age, years Male:Female Headache Visual complaints Papilledema Multiple cranial nerve involvement Ataxia Seizures Hemiparesis CSF protein, mg% CSF pleocytosis, mean MRI: dural enhancement Dural biopsy: pachymeningitis Steroid use Prognosis
12 55 (Range: 39–88) 9:3 11 7 2 4
14 69 (Range: 37–88) 8:6 9 9
20 49 (Range: 30–67) 11:9 16 15 8 10
2 1 – 81(Range: 30–243) 4 12 5 12 Seven patients had significant improvement, headache & ataxia persisted in one each, one had memory dysfunction and two expired
1 1 1 190(Range: 17–1654) 20 14 6 13 Seven patients had relapse of symptoms as the corticosteroid dose was tapered off. Two deaths (one unrelated).
added [18]. One case showed improvement with chloroquine and radiotherapy only. But the abnormal dural enhancement in MRI can persist despite a good clinical response to corticosteroids or other immunosuppressive agents or both as in 13(65%) of our patients with IHP. Except one, all our patients were on immunomodulatory treatment during the last follow-up. In short, these patients exhibit a very good clinical improvement with steroids alone in majority, especially the most common clinical symptoms and signs pertaining to optic and oculomotor nerves respond well to steroids. The other immunomodulatory drugs may have a role as a steroid sparing agent only. There are no clear guidelines as to the specific dosage or duration of therapy with steroids; one can possibly conclude that 1–2 mg/Kg of prednisolone for a minimum period of 12 weeks or may be extended for more days or weeks based on the treatment response. One has to know that immunomodulation should not be continued or incremented based on the MRI response since in IHP, the MRI seldom returns to normalcy as one expects to happen with the clinical improvement as discussed earlier. The number of follow-up MRIs needed after initiation of treatment also should be guided by clinical improvement in IHP. Majority of our patients showed a significant improvement in the functional scale after successful treatment. These simple scales could be applied meaningfully by chart review of respective patients from the prospectively maintained records in our institute. We do not want to conclude this scale as a validated one which is practically not possible in this disease which is by and large a heterogeneous entity at presentation and in later follow-up. However, the scale is put forth only to incorporate all the symptoms and signs which can occur on neurological examination in the order of commonality and prevalence in pachymeningitis. By doing so, one can have a uniform evaluation of an individual patient during his/her follow-up. For example person presenting with headache alone can be made to visually and subjectively score his only subjective symptom before start of the treatment and at various periods of follow-up to decide on improvement/worsening and future treatment plan since there are no uniform guidelines for treatment existing in literature and since the radiological examination may not help always. After accruing a good number of patients prospectively, we are planning to validate this scale if possible which may then give more meaningful information. But this may be considered a pragmatic, preliminary tool towards reaching that stage. Such
6
5 6 2 96(Range: 30–494) 8 20 9 20 Headache recovered in 13 patients, improved but not recovered fully in another 7 patients. Cranial nerve palsies recovered in all but three. Significant difference between the pretreatment and post-treatment scores
simple scales helps the treating physician to follow-up the patients more easily and assess the responsiveness to treatment. We acknowledge the following limitations of our study. First of all due to financial constraints and lack of availability in a developing nation, we were not able to measure the CSF and serum IgG index in all cases of HP. Further, the lack of a large sample size in our study might have resulted in bias in the interpretation of the MRI data to assess the real value of “Eiffel-by-night sign” in distinguishing idiopathic from secondary causes of pachymeningitis. However, it remains a novel MRI finding and can be elucidated more clearly by analyzing cases worldover by a systematic review. 5. Conclusions To the best of our knowledge, this is the largest series of IHP described so far. Though biopsy is the gold standard for definitive diagnosis of IHP to differentiate it from other causes, it may not always be necessary or may be feasible only in higher centers with ample facilities. Hence, a clinician should largely depend on the clinical profile, extensive laboratory investigations to exclude other causes of pachymeningitis and above all, the MRI findings. We have highlighted the role of MRI in differentiating between IHP and other causes of SHP. Competing interests/conflict of interest Nil. Funding source Nil. Disclosure The authors report no disclosures relevant to the manuscript. Acknowledgments We acknowledge Dr. Ravi Prasad Varma for his help in statistical analysis.
58
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
Appendix 1. Etiology of hypertrophic pachymeningitis
Primary or idiopathic hypertrophic pachymeningitis Secondary causes Infectious diseases Mycobacterium tuberculosis Fungal infections Lyme disease Syphilis Eosinophilic meningitis due to parasitic diseases Systemic autoimmune/vasculitic disorders Sarcoidosis Wegener's granulomatosis Rheumatoid arthritis Behcet's disease Sjogren syndrome Temporal arteritis Mixed connective tissue disorder Orbital pseudotumor Vogt–Koyanagi–Harada disease IgG4 related inflammatory meningeal diseases Inflammatory bowel disease Others Dural carcinomatosis Meningioma Metastatic disease in adjacent skull bone Meningeal lymphoma POEMS (Polyneuropathy, Organomegaly, Endocrinopathy, Monoclonal plasma proliferative disorder, skin changes) Medications: Non-steroidal anti-inflammatory agents
Appendix 2a. Clinical scores of patients with hypertrophic pachymeningitis*
Headache 0 = no headache 1 = mild headache 2 = moderate 3 = severe 4 = very severe, disabled, confined to bed Encephalopathy 0 = absence of either seizure/encephalopathy 1 = presence of either encephalopathy/seizures Visual symptoms 0 = no visual symptoms 1 = present, not interfering with daily functioning 2 = present and interfering with daily functioning Ataxia 0 = no ataxia 1 = mild, able to move independently 2 = moderate, need 1 person's support 3 = severe, confined to bed/chair Hemiparesis/Quadriparesis 0 = no hemiparesis 1 = mild, able to move independently 2 = moderate, need 1 person's support 3 = severe, confined to bed/chair Other cranial nerve dysfunction** 0 = no symptoms 1 = present, not interfering with daily functioning 2 = present and interfering with daily functioning *Each symptom reported to be associated with pachymeningitis were scored separately and then the sum of the score was taken as the “composite score”. **For each cranial nerve dysfunction, the maximum score is 2 and the minimum score is 0 (except the optic and oculomotor nerves and the lower cranial nerves i.e. 9th, 10th, 11th & 12th nerves which are given scores taking them as a group) and thus the maximum total score for cranial nerve dysfunction is 12 and the maximum total clinical score is 23.
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
Appendix 2b. Clinical scores of patients with hypertrophic pachymeningitis
59
60
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
Appendix 3. The variability of clinical scores between clinician and patients
Variable
Kappa*
Level of agreement
Headache Encephalopathy Visual score Ataxia Hemiparesis Bulbar palsy Other cranial nerve score
0.529 1.0 0.836 1.0 1.0 1.0 0.906
Moderate Almost perfect Almost perfect Almost perfect Almost perfect Almost perfect Almost perfect
*Reference: Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33 [1]: 159–174.
References [1] Mamelak AN, Kelly WM, Davis RL, et al. Idiopathic hypertrophic cranial pachymeningitis. Report of 3 cases. J Neurosurg 1993;79:270–6. [2] Masson C, Hénin D, Hauw JJ, et al. Cranial pachymeningitis of unknown cause. A study of seven cases. Neurology 1993;43:1329–34. [3] Naffziger HC, Stern WE. Chronic pachymeningitis: report of a case and review of the literature. Arch Neuro Psychiatry 1949;62:383–411. [4] Parker Stephen W, Sobel Raymond A. Case 12—1988. N Engl J Med 1988;318:760–8. [5] Schwartz William J, Niles John L. Case 9—1999. N Engl J Med 1999;340:945–54. [6] Carruthers R, Carruthers M, Della-Torre E. IgG4-related disease and other causes of inflammatory meningeal disease. Semin Neurol 2014;34:395–404. [7] Lu LX, Della-Torre E, Stone JH, et al. IgG4-related hypertrophic pachymeningitis: clinical features, diagnostic criteria, and treatment. JAMA Neurol 2014;71:785–93.8. [8] Bonita R, Beaglehole R. Recovery of motor function after stroke. Stroke 1988;19: 1497–500. [9] Swieten JC, Koudstaal PJ, Visser MC, et al. Interobserver agreement for the assessment of handicap in stroke patients. Stroke 1988;19:604–7. [10] Stewart WF, Lipton RB, Kolodner K. Migraine disability assessment (MIDAS) score: relation to headache frequency, pain intensity, and headache symptoms. Headache 2003;43:258–65.
[11] Schmahmann JD, Gardner R, MacMore J, et al. Development of a brief ataxia rating scale (BARS) based on a modified form of the ICARS. Mov Disord 2009;24:1820–8. [12] Kupersmith MJ, Martin V, Heller G, et al. Idiopathic hypertrophic pachymeningitis. Neurology 2004;62:686–94. [13] Hatano N, Behari S, Nagatani T, et al. Idiopathic hypertrophic cranial pachymeningitis: clinicoradiological spectrum and therapeutic options. Neurosurgery 1999;45:1336–44. [14] Thomas B, Thamburaj K, Kesavadas C. ‘Eiffel-by-Night’: a new MR sign demonstrating reactivation in idiopathic hypertrophic pachymeningitis. Neuroradiol J 2007;20: 194–5. [15] Riku S, Kato S. Idiopathic hypertrophic pachymeningitis. Neuropathology 2003;23: 335–44. [16] Sylaja PN, Cherian PJ, Das CK, et al. Idiopathic hypertrophic cranial pachymeningitis. Neurol India 2002;50:53–9. [17] Della-Torre Galli L, Franciotta D, et al. Diagnostic value of IgG4 Indices in IgG4related hypertrophic pachymeningitis. J Neuroimmunol 2014:266:82–6. [18] Hamilton S, Smith C, Lessell S. Idiopathic hypertrophic cranial pachymeningitis. J Clin Neuroophthalmol 1993;13:127–34.
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Clinico-radiological spectrum and outcome in idiopathic hypertrophic pachymeningitis Gopal Krishna Dash, Bejoy Thomas, Muralidharan Nair, Ashalatha Radhakrishnan ⁎ Department of Neurology and Imaging & Interventional Radiology, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Trivandrum, India
a r t i c l e
i n f o
Article history: Received 31 October 2014 Received in revised form 25 January 2015 Accepted 5 February 2015 Available online 14 February 2015 Keywords: Hypertrophic pachymeningitis Clinical features Radiological features Treatment modalities Outcome
a b s t r a c t Objective: To elucidate the clinico-radiological features, treatment response and outcome of a large cohort of patients (n = 20) with idiopathic hypertrophic pachymeningitis (IHP) and to examine if any of these features could differentiate between IHP and secondary causes of hypertrophic pachymeningitis (SHP). Methods: 20 patients with IHP diagnosed between 1998 and 2009 formed the study cohort. We adopted a validated clinical score to quantitatively assess and document their neurological disability and to compare their pre- and post-treatment outcomes. Appropriate statistical analysis was done to look for any clinical and/or radiological features to differentiate IHP from SHP. Results: Out of the twenty eight consecutive patients with pachymeningitis, 20 were having IHP and 8 were having SHP (Tuberculosis-5, Sarcoidosis-2, Wegener's granulomatosis-1). In IHP, headache and visual symptoms dominated the clinical symptomatology (80% and 75%). In MRI, the peripheral pattern of contrast enhancement was more common with IHP (p = 0.03). The posterior falx and tentorium showing a hypointense center (“fibrosis”) and enhancing periphery (“active inflammation”) together mimicking “Eiffel-by-night” sign was found to be more commonly associated with IHP (60% vs 12.5%, p = 0.03). Biopsy was done in 9 patients. At a mean follow-up of 51 months (range 24–144 months), the mean pretreatment clinical score improved from 6.55 to 1.80 in 20 patients with IHP (p b 0.001). Conclusions: Our data on the largest cohort of patients with IHP would shed light into its clinico-radiological spectrum, treatment and outcome. The prognosis is satisfactory if managed appropriately. We have highlighted the role of MRI in differentiating between IHP and other causes of SHP. © 2015 Published by Elsevier B.V.
1. Introduction Hypertrophic pachymeningitis (HP) is characterized by localized or diffuse thickening of the cranial or spinal dura mater which may or may not be associated with inflammation, resulting in progressive neurological deficits [1,2]. It has been described in association with infection, trauma, tumors, and Wegener's granulomatosis [3–5] (Appendix 1). IgG4-related hypertrophic pachymeningitis (IgG4-RHP), a recently described entity, is an increasingly recognized manifestation of IgG4-related disease, a fibroinflammatory condition that can affect virtually any organ. It is estimated that IgG4-RHP may account for a high proportion of cases of hypertrophic pachymeningitis once considered idiopathic [6,7].
⁎ Corresponding author at: Department of Neurology, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Trivandrum, Kerala, India-695011. Tel.: + 91 471 2524684; fax: +91 471 2446433. E-mail address: [email protected] (A. Radhakrishnan).
http://dx.doi.org/10.1016/j.jns.2015.02.013 0022-510X/© 2015 Published by Elsevier B.V.
From the management point of view, attempts to differentiate primary or idiopathic HP (IHP) from secondary hypertrophic pachymeningitis (SHP) is very important. Definite diagnosis can be obtained after meningeal biopsy, but may not always be possible or warranted. MRI characterizes the degree of dural inflammation and clinches the diagnosis of HP, but till date there is no data on how the imaging helps in differentiating IHP and SHP. Most available literature on IHP being either case reports or small series, focusing on MRI abnormalities or the variable clinical manifestations and the clinical outcome with short follow-up, one cannot deduce any conclusive data on the same. No objective scales assessing the disability have ever been used to quantify the long-term functional outcome of a large group of IHP patients. This is especially important because of the prevailing notion that a majority of these patients do not recover fully and it is a long-standing, chronic disease with remission and relapses. Therefore, we attempted to elucidate the clinico-radiological and laboratory features, treatment response and outcome of a large cohort of patients with pachymeningitis. We specifically examined if any of these features could distinguish IHP from SHP thus aiding the treating neurologist.
52
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
2. Methods
2.4. Brain biopsy
2.1. Subjects and methodology
Biopsy was performed in 9 patients and included meninges and/or brain parenchyma. They were stained for infectious agents and cultures were done.
This study was conducted at the Sree Chitra Tirunal Institute for Medical Sciences and Technology, a tertiary center for neurological disorders in India. Institutional ethics committee approval was obtained for the study. 28 consecutive patients with hypertrophic pachymeningitis diagnosed between January 1998 and December 2009 were evaluated in detail. We adopted a simple composite clinical score derived from the various clinical manifestations in patients with hypertrophic pachymeningitis available in published literature and then compared the pre- and post-treatment score to evaluate their outcome homogeneously. Each component of the score has been adopted and modified from previously validated clinical scores [8–11]. The details of the clinical scores are provided in Appendix 2a. The authors (GKD and AR) did a detailed chart review to score each of the patient's status at various periods of follow-up, this being a simple scale and the scoring could easily be accomplished in all patients. Each patient was then personally interviewed by the Neurologists (GKD and AR) and patients were also made to score their symptoms before and after treatment in a Likert scale independently (Appendix 2b). The variability if any, in each of the score pertaining to a specific symptom and the composite score as scored by the authors and the patients were statistically analyzed by kappa statistics.
2.2. Investigations All patients underwent the following investigations: hemogram (hemoglobin, differential and total leukocyte counts, total platelet count), erythrocyte sedimentation rate (ESR), biochemical investigations including serum calcium, phosphorus, liver function tests, renal function tests, serology for human immunodeficiency virus (HIV), rheumatoid factor(RA factor), antinuclear antibodies, APLA (antiphospholipid antibody),serum VDRL (venereal disease reference laboratory testing),P and C antineutrophil cytoplasmic antibodies (cANCA and pANCA) and angiotensin converting enzyme(ACE) assays. Cerebrospinal fluid (CSF) examination was done in all patients for cytology, biochemistry and immunological tests to exclude secondary causes. A Mantoux test was performed in all patients.
2.3. Imaging Imaging studies included chest X-ray and contrast enhanced MRI brain (1.5 T Scanner, Signa GE, Milwaukee, WI) in all patients. All the MRI scans archived in our system were analyzed by the neuroradiologist (BT) blinded to the clinical data. Follow-up contrast enhanced MRI was performed if clinically indicated, and the findings were classified as (i) improved, (ii) worsened, or (iii) no change in relation to the immediately prior MRI(s). In the final follow-up MRI, the degree of abnormal enhancement was defined as (i) stable when unchanged, (ii) improved when there was less extensive enhancement or reduced thickness of the dura mater, and (iii) worse if there was increased enhancement compared to the initial imaging. [12] The patterns and sites of enhancement were noted and characterized as described by Hatano et al. in 1999 [13]. It was described as linear, nodular and combined. We also classified the pattern of dural enhancement as “peripheral”, “uniform” or “combined”. The involvement of pachymeninges was described as diffuse if more than 2 non-contiguous sites were involved and focal if less than 2 sites were involved and were also subclassified as “symmetric” or “asymmetric” by comparing either side of the brain.
2.5. Statistical analysis All analyses were performed using SPSS version 17.0 (SPSS Inc., Chicago, IL). Fisher's exact test, Paired t-test, kappa statistics, positive and negative predictive values, sensitivity and specificity were applied as appropriate. A p-value ≤0.05 was taken as significant. 3. Results 3.1. Clinical profile Out of 28 patients with HP, 20 patients had IHP and 8 patients had SHP. In the IHP group, there were 11 men and 9 women with a median
Table 1 Clinical features, CSF and MRI findings in patients with hypertrophic pachymeningitis (N = 28). Characteristics
IHP (N = 20)
SHP (N = 8)
Significance p*
Age, mean, years Sex (female:male) Age at onset of symptoms, mean, years Duration of illness, mean, years Clinical features Headache Visual symptoms Seizure Hearing loss Dysphagia Dysarthria Ataxia Facial sensory complaints Hemiparesis Quadriparesis Only second cranial nerve involvement Second cranial nerve involvement with other cranial nerve palsies Involvement of oculomotor nerves Facial nerve involvement Eight cranial nerve involvement Bulbar palsy CSF protein, mg% (mean, range) CSF glucose, mg% (mean, range) CSF cell count (mean, range) Site(s) of involvement in MRI Falcotentorial Medial frontal Basifrontal Frontal convexity Temporal convexity Skull base Symmetrical involvement in MRI Diffuse involvement Spinal involvement T2 Signal in MRI Hypointense Isointense Peripheral pattern of CE
49.5 9:11 44.5
40.8 5:3 39.7
NS NS NS
5.1 N, % 16(80.0) 15(75.0) 6(30.0) 4(20.0) 3(15.0) 2(10.0) 5(25.0) 2(10.0) 2(10.0) 2(10.0) 5(25.0)
1.28 N, % 7(87.5) 3(37.5) 2(25.0) 2(25.0) 1(12.5) 1(12.5) 2(25.0) 1(12.5) 0(0.0) 0(0.0) 2(25.0)
NS NS NS NS NS NS NS NS NS NS NS NS
5(25.0)
1(12.5)
NS
7(35.0) 2(10.0) 4(20.0) 5(25.0) 96.9(30–494) 82(50–154) 35(2–460) I 17(85.0) 7(35.0) 7(35.0) 6(30.0) 6(30.0) 5(25.0) 10(50.0) 13(65.0) 2(10)
3(37.5) 0(0.0) 2(25.0) 3(37.5) 76(37–159) 82.8(53–153) 16(2–75)
NS NS NS NS NS NS NS
4(50.0) 6(75.0) 4(50.0) 5(62.5) 3(37.5) 2(25.0) 4(50.0) 7(87.5) 0(0.0)
NS NS NS NS NS NS NS NS NS
19(95.0) 1(5.0) 12(60.0) 12(60.0) 3(15.0)
6(75.0) 2(25.0) 1(12.5) 1(12.5) 5(62.5)
NS NS NS 0.03 0.02
Follow-up MRI showing improvement
IHP — idiopathic hypertrophic pachymeningitis, SHP — secondary hypertrophic pachymeningitis, N — number, NS — not significant, * — p-value by Fisher's exact test, ^ — see text for description, CE — contrast enhancement.
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
53
Fig. 1. Graphical representation of clinical features in patients with idiopathic hypertrophic pachymeningitis in percentages.
age at presentation of 48 years (range 50–80 years). The details of their clinical features are provided in Table 1 and Fig. 1. The most common presentation in patients with IHP was headache (16 / 20, 80%). Poor visual acuity was noted in 9 patients with IHP, which ranged from 6/9 (Snellen's chart) to perception of light only. The most common cranial nerve involved in isolation was the second cranial nerve, seen in 5(25%) patients. No clinical feature could differentiate IHP from SHP (Table 1). At a mean follow-up of 51 months (range 24–144 months), the mean clinical score improved from 6.55(pre-treatment) to 1.80(post-treatment) in patients with IHP (p b 0.001), Fig. 2. The variability in clinical scoring between the patient and clinician remained “almost perfect” (Kappa varied between 0.8 and 1), except that in the severity of headache, the score showed only a “moderate agreement” (Kappa = 0.529) (Appendix 3). 3.2. CSF and other laboratory findings The CSF findings are summarized in Table 1. CSF parameters could not differentiate between IHP and SHP. In the IHP group, the mean ESR was 49 mm/h (range 4–124). RA factor and APLA were
positive in one patient each. CT scan of thorax and abdomen was normal in all patients with a diagnosis of IHP. In 4 patients with SHP, Mantoux test was positive and one patient had cANCA positivity.
3.3. Neuroimaging findings The details of the MRI findings in patients with HP are summarized in Table 1. In the IHP group, the most common pattern of pachymeningeal involvement was falcotentorial, seen in 17(85%) patients and the commonest pattern of contrast enhancement was a peripheral pattern (12, 55%) followed by a uniform pattern. Combined pattern (both peripheral and uniform) of enhancement was noted in only one patient, while nodular pattern was not seen. Associated spinal involvement was seen in 2 patients with IHP and none in those with SHP. Associated brain parenchymal involvement was noted in 6 patients (Fig. 3), venous compression in 6, leptomeningeal involvement in 3 and venous thrombosis in one patient. Significantly more number of patients with IHP had peripheral pattern of contrast enhancement (12/20 vs 1/8, p = 0.03) and they exhibited “Eiffel-by-night” sign [14] in MRI (12 vs 1, p = 0.03). This sign was described by our group earlier and it represents thickened dura in the posterior falx and tentorium due to fibrosis seen as central hypointensity in post-contrast fat saturated T1W image with an enhancing periphery depicting areas of active inflammation, mimicking the illuminated Eiffel tower at night (Fig. 4). The positive and negative predictive values, sensitivity and specificity of this sign in predicting the diagnosis of IHP were 92%, 46%, 60% and 87% respectively. Follow-up scans were available in 13 patients with IHP, the findings remained unchanged in 11 and showed progression in 2 patients during follow-up. More number of patients with SHP had improvement in their MRI scan as opposed to IHP group (3 vs 5, p = 0.02) (Fig. 5).
3.4. Histopathological findings
Fig. 2. Box plot depicting the total clinical scores of 20 patients with idiopathic hypertrophic pachymeningitis before and after treatment. Solid horizontal lines are median values, and length boxes are interquartile ranges (25th to 75th percentiles) excluding the outliers. The tip of the vertical lines from the top and bottom of the boxes indicates the maximum and minimum values excluding the extreme values (*).
Histopathological data was available in 9 IHP patients who underwent biopsy. Meningeal biopsies revealed dense fibrous infiltration with few plasma cells and lymphocytes. There was no granuloma formation, vasculitis or neoplastic cells (Fig. 6, patient 18). Stains for fungus, acid fast bacilli and cultures were negative in all cases of IHP.
54
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
Fig. 3. A) Axial fluid attenuated inversion recovery sequence image (FLAIR) of patient 8 showing significant bilateral occipital hyperintensity suggestive of edema. B) T1 weighted axial post contrast fat sat image showing contrast enhancement in the posterior falx (arrow). C) Axial FLAIR image of the same patient after one year of treatment showing significant resolution of edema and D) decrease in contrast enhancement.
3.5. Treatment and outcome The most common drug used for treatment in IHP was prednisolone. Other immunomodulatory agents used were azathioprine in 3, methotrexate in one, cyclophosphamide in one and mycophenolate mofetil in one patient respectively. During a mean follow-up of 51 months (range 24–144 months), headache recovered in 13(65%) patients and improved but not recovered fully in another 7(35%) patients. Except in 3(15%) patients, cranial nerve palsies recovered in all. There was a significant difference between the pretreatment and post-treatment score (6.55 vs 1.80, p b 0.001) (Fig. 2). In patients with SHP, 5 were treated with antitubercular drugs, one with prednisolone and two received combined prednisolone
and azathioprine(for sarcoidosis and Wegener's granulomatosis respectively). We have summarized the data of all 20 cases of IHP in Table 2. We have also compared the various large series of patients with IHP reported so far with ours in Table 3. 4. Discussion In this largest cohort of 20 patients with IHP, many patients were in their late fifties during presentation similar to previously reported series [12,15]. Both the genders were affected equally [2, 13]. In patients with IHP, headache was the most common presentation. Other small series are also in agreement with this observation [2,12,16]. Headache is due mainly to inflammation of the dura
Fig. 4. (A) T2-W coronal image at the falco-tentorial junction of patient 10 demonstrates the thick hypointense dura. (B) Contrast enhanced T1-W fat saturated coronal image through the same region shows thickened dura with hypointense center and enhancing periphery. (C) Contrast enhanced image 3 years later in the same patient shows persistence of contrast enhancement. (D) The peripheral enhancement pattern of the thick hypointense dura mimics the image of illuminated Eiffel tower (‘Eiffel-by-night’ sign).
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
55
Fig. 5. (A) Contrast enhanced fat saturated T1-W coronal image through the posterior falx and tentorium shows uniform enhancement in the tentorium, in a patient with secondary hypertrophic pachymeningitis due to tuberculosis (‘Eiffel-by-night’ sign negative). (B) Contrast enhanced image two years later in the same patient shows significant improvement of the dural enhancement.
mater, but can also be due to raised intracranial pressure [12,15]. The next most common manifestation was visual loss due to optic neuropathy, subsequent optic atrophy and the oculomotor nerve (i.e. third, fourth, and sixth cranial nerves) involvement. This is in contrast to a previously reported series comprising 12 patients who were the largest cohort described so far, where visual loss was seen in 11 out of 12 patients [12]. Such discrepancies are expected in rare diseases since no single center can have a large group of patients to audit the manifestations accurately. Apart from the second and oculomotor nerve involvement, bulbar palsy was noted in 25%, seventh and eighth cranial nerve involvement in 10 and 20% respectively. Some studies on IHP have described seventh nerve followed by lower cranial nerves as the most common pattern of cranial neuropathy [12]. However, we found that the involvement of optic and oculomotor nerves was more common. Seizures occurred in 30% and encephalopathy and hemiparesis in 20% each. Seizures albeit rare, can occur in patients with IHP related to parenchymal edema adjacent to the inflamed meninges or venous infarct. None of the clinical findings or routine laboratory features could
Fig. 6. Histopathology in IHP (patient 18) (hematoxylin and eosin stain ×500) showing the dura with dense fibrocollagenous tissue only.
differentiate IHP from SHP. This underscores the importance of a thorough evaluation by specific laboratory investigations, MRI and if needed biopsy in all patients with pachymeningitis since no clinical or laboratory parameter can sufficiently bolster ones clinical certainty as to the cause of HP. Contrast MRI of the brain and spinal cord is one of the key investigations in patients with HP which demonstrates the thickened and abnormally enhancing dura mater. It can assess the localization, extent, and pattern of involvement of abnormal dura and helps to monitor the progression. The sites of abnormal dural enhancement most commonly seen in patients with IHP were falco-tentorial (85%), medial and basifrontal (35%) followed by frontal and temporal convexities and skull base. We did not find a significant difference between the sites of involvement in IHP and SHP. A similar pattern of dural enhancement was noted in a previous series by Kupersmith et al [12]. However, in contrast to their series, we did not observe a dominant involvement of cavernous sinus. The authors have pointed out that the abnormal enhancement of sphenoid wing dura mater and cavernous sinus (100%) in their series might reflect a bias to patients with neuro-ophthalmologic symptoms since there was an ophthalmic center. We observed both symmetric and asymmetric pattern of dural enhancement occurring equally in our cohort (50% each), but more patients had a peripheral pattern of enhancement (60%). ‘Eiffel-by-night’ sign, described previously [14] by our group was noted in 60% of IHP patients, while only 12% with SHP had a similar MRI enhancement (p-0.03). In IHP, pathologically diffuse thickening of the dura with dense fibrosis and inflammatory cell infiltration more marked at the periphery composed of lymphocytes and plasmacytes was noted. This fibrosis imparts a hypointense signal on all sequences in MRI and the site of active inflammation on the surface is marked by contrast enhancement. This new sign encompassing all these features should therefore be specifically looked for in all cases of IHP and if found points towards IHP. We could not account for the pathogenesis of this specific pattern of involvement in IHP. Similarly, significant improvement in the degree of pachymeningeal thickening in MRI after treatment occurs only in secondary causes of HP, whereas in IHP, even with clinical improvement the same degree of MRI improvement in the form of reduction in the thickening of dura does not occur. At a median follow-up of 24 months, despite clinical improvement, 8 out of 20 patients continued to have similar imaging findings in IHP. Hence one should not be over enthusiastic in treating such cases with immunomodulators
56
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
Table 2 Summary of clinical, CSF, MRI findings, treatment and outcome of patients with IHP. Pt Age/Sex Clinical presentation No 1
51/F
2
53/M
3
40/M
4
64/M
5
57/M
6
43/F
7
50/M
8 9
54/F 30/M
10
35/M
11
39/F
12
67/M
13
38/F
14 15
49/F 37/F
16
47/M
17
61/M
18
38/F
19
80/M
20
58/F
Pre-treatment CSF clinical score* protein (mg/dl)
Headache and 3rd, 5th, and 6th CN 5 involvement Headache, seizures, and 2nd CN 5 involvement Headache, hemiparesis, and 2nd CN 7 involvement Headache, ataxia, and 2nd 8th CN 7 involvement Headache, seizure, and 2nd & 6th CN 7 involvement Headache, encephalopathy, and 2nd, 9 3rd, 6th, and 8th CN involvement Ataxia, hemiparesis 2nd, and 8th CN 5 involvement Headache and encephalopathy 2 Headache, seizure 2nd CN 7 involvement Headache, and 2nd, 3rd, & 6th CN 6 involvement Seizure, and 2nd & 3rd CN 5 involvement Headache, and 2nd, 3rd, 5th, 7th, 9th, 8 10th, and 11th CN involvement 6 Headache, ataxia, hemiparesis, and 2nd, 6th, 9th, 10th, 11th, and 12th CN involvement Headache, and 2nd CN involvement 10 Headache, seizures, and 6th CN 5 involvement Headache, and 2nd, 9th, 10th, 11th, 4 and 12th CN involvement Ataxia, hemiparesis, and 8th CN 6 involvement Headache, seizures, and 2nd CN 5 involvement Headache, and 2nd, 5th, and 6th CN 3 involvement Headache, and 2nd, 8th, 9th, 10th, and 11 11th CN involvement
CSF MRI* (Pattern of cells pachymeningeal thickening)
Eiffel-by-night Biopsy Treatment Follow-up sign* clinical score*
Follow-up duration, months
87
2
Diffuse, symmetric
+
No
Pred
1
24
32
2
Diffuse, symmetric
+
No
Pred
0
48
56
6
Focal, asymmetric
−
Yes
Pred
0
48
359
460
Focal, asymmetric
−
No
Pred
2
48
109
50
Diffuse, symmetric
+
Yes
Pred
0
18
42
2
Focal, asymmetric
−
Yes
Pred
5
144
84
40
Diffuse, symmetric
+
No
Pred
1
18
26 55
2 2
Diffuse, symmetric Focal, asymmetric
+ +
No No
Pred, Myc Pred
0 0
24 26
87
5
Diffuse, symmetric
+
Yes
Pred
3
24
59
3
Diffuse, asymmetric
+
No
Pred
0
40
54
5
Focal, asymmetric
−
No
Pred, Az
1
24
56
50
Focal, asymmetric
−
Yes
Pred
2
84
53 45
10 2
Focal, asymmetric Diffuse, symmetric
− +
Yes No
Pred Pred, Mtx
3 1
25 84
44
10
Diffuse, asymmetric
+
Yes
Pred, Az
1
84
494
20
Focal, asymmetric
−
No
Pred, Az
2
18
71
5
Diffuse, symmetric
+
Yes
Pred
2
18
55
10
Diffuse, symmetric
+
No
Pred
1
18
30
2
Diffuse, symmetric
+
Yes
Pred
0
14
IHP — idiopathic hypertrophic pachymeningitis, M — male, F — female, CN — cranial nerve, Pred — prednisolone, Az — azathioprine, Myc — mycophenolate mofetil, Mtx — methotrexate, * — see text for definition.
repeatedly even if the MRI is not showing an improvement pari passu with the clinical. Although biopsy is considered the gold standard for the definitive diagnosis of HP, the indications for biopsy is still not clearly laid down due to scarcity of literature. One can plausibly suggest that biopsy of the dural lesion should be considered if the patient diagnosed as IHP clinically deteriorates or the neuroimaging worsens despite treatment when the investigations fail to detect any secondary cause for HP. Chronic inflammation of the usually inert dura mater with minimal or no inflammation of the pia and arachnoid is seen in IHP. Lymphocytes, plasma cells, macrophages, epithelioid cells, and occasionally granulomas infiltrate the dura, later to be replaced by fibrosis. In the series by Kupersmith et al. on 12, two developed malignancy (one patient had carcinoma breast and the other had prostate malignancy) within a month of diagnosis of IHP. But such changes if destined to happen would become evident within few months and therefore a stringent follow-up of a minimum of one year is warranted. IgG4-related disease, a recently described condition with inflammatory meningeal disease accounts for an increasing proportion of cases of idiopathic hypertrophic pachymeningitis. Clinicians should become familiar with this alternative differential diagnosis and should appropriately test for the CSF and serum for the presence of immunoglobulin [6,7]. In addition to the MRI findings, CSF analysis, particularly CSF IgG (specifically
IgG4 levels) index and serum IgG analysis will unfold the etiological diagnosis of patients with inflammatory meningeal diseases [17]. These tests should be included while evaluating patients with hypertrophic pachymeningitis. Patients with IHP frequently progresses, if not managed properly. Steroid was the preferred drug in all our patients. Azathioprine, methotrexate and mycophenolate were used in few patients additionally (5 patients). During a mean follow-up of 51 months (range 24–144 months), headache recovered in 13 patients (65%) and improved, but not recovered fully in another 7(35%) patients. Except in 3(15%), cranial nerve palsies recovered in all patients. In a previous series of IHP by Masson et al., all 7 patients had relapse of symptoms after withdrawing the steroids [2]. Out of 12 patients in Kupersmith series, 7 out of 8 patients with visual loss improved rapidly and maintained their vision on therapy, suggesting that visual loss could be secondary to inflammation rather than the compressive fibrosis around the optic nerve [12]. Similarly, the gait imbalance, hemiparesis, and epilepsy also improved with corticosteroid treatment. Previous studies have shown that azathioprine or methotrexate allows tapering of steroids [12,10]. Hamilton et al., described 3 patients with optic neuropathy or painful ophthalmoparesis or both where the symptoms in two cases were controlled only by high dose prednisolone, until azathioprine was
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
57
Table 3 Comparison of the salient features in three large studies of hypertrophic cranial pachymeningitis. Parameters
Kupersmith (2004)
Riku S (2003)
Present study
Number of patients Mean age, years Male:Female Headache Visual complaints Papilledema Multiple cranial nerve involvement Ataxia Seizures Hemiparesis CSF protein, mg% CSF pleocytosis, mean MRI: dural enhancement Dural biopsy: pachymeningitis Steroid use Prognosis
12 55 (Range: 39–88) 9:3 11 7 2 4
14 69 (Range: 37–88) 8:6 9 9
20 49 (Range: 30–67) 11:9 16 15 8 10
2 1 – 81(Range: 30–243) 4 12 5 12 Seven patients had significant improvement, headache & ataxia persisted in one each, one had memory dysfunction and two expired
1 1 1 190(Range: 17–1654) 20 14 6 13 Seven patients had relapse of symptoms as the corticosteroid dose was tapered off. Two deaths (one unrelated).
added [18]. One case showed improvement with chloroquine and radiotherapy only. But the abnormal dural enhancement in MRI can persist despite a good clinical response to corticosteroids or other immunosuppressive agents or both as in 13(65%) of our patients with IHP. Except one, all our patients were on immunomodulatory treatment during the last follow-up. In short, these patients exhibit a very good clinical improvement with steroids alone in majority, especially the most common clinical symptoms and signs pertaining to optic and oculomotor nerves respond well to steroids. The other immunomodulatory drugs may have a role as a steroid sparing agent only. There are no clear guidelines as to the specific dosage or duration of therapy with steroids; one can possibly conclude that 1–2 mg/Kg of prednisolone for a minimum period of 12 weeks or may be extended for more days or weeks based on the treatment response. One has to know that immunomodulation should not be continued or incremented based on the MRI response since in IHP, the MRI seldom returns to normalcy as one expects to happen with the clinical improvement as discussed earlier. The number of follow-up MRIs needed after initiation of treatment also should be guided by clinical improvement in IHP. Majority of our patients showed a significant improvement in the functional scale after successful treatment. These simple scales could be applied meaningfully by chart review of respective patients from the prospectively maintained records in our institute. We do not want to conclude this scale as a validated one which is practically not possible in this disease which is by and large a heterogeneous entity at presentation and in later follow-up. However, the scale is put forth only to incorporate all the symptoms and signs which can occur on neurological examination in the order of commonality and prevalence in pachymeningitis. By doing so, one can have a uniform evaluation of an individual patient during his/her follow-up. For example person presenting with headache alone can be made to visually and subjectively score his only subjective symptom before start of the treatment and at various periods of follow-up to decide on improvement/worsening and future treatment plan since there are no uniform guidelines for treatment existing in literature and since the radiological examination may not help always. After accruing a good number of patients prospectively, we are planning to validate this scale if possible which may then give more meaningful information. But this may be considered a pragmatic, preliminary tool towards reaching that stage. Such
6
5 6 2 96(Range: 30–494) 8 20 9 20 Headache recovered in 13 patients, improved but not recovered fully in another 7 patients. Cranial nerve palsies recovered in all but three. Significant difference between the pretreatment and post-treatment scores
simple scales helps the treating physician to follow-up the patients more easily and assess the responsiveness to treatment. We acknowledge the following limitations of our study. First of all due to financial constraints and lack of availability in a developing nation, we were not able to measure the CSF and serum IgG index in all cases of HP. Further, the lack of a large sample size in our study might have resulted in bias in the interpretation of the MRI data to assess the real value of “Eiffel-by-night sign” in distinguishing idiopathic from secondary causes of pachymeningitis. However, it remains a novel MRI finding and can be elucidated more clearly by analyzing cases worldover by a systematic review. 5. Conclusions To the best of our knowledge, this is the largest series of IHP described so far. Though biopsy is the gold standard for definitive diagnosis of IHP to differentiate it from other causes, it may not always be necessary or may be feasible only in higher centers with ample facilities. Hence, a clinician should largely depend on the clinical profile, extensive laboratory investigations to exclude other causes of pachymeningitis and above all, the MRI findings. We have highlighted the role of MRI in differentiating between IHP and other causes of SHP. Competing interests/conflict of interest Nil. Funding source Nil. Disclosure The authors report no disclosures relevant to the manuscript. Acknowledgments We acknowledge Dr. Ravi Prasad Varma for his help in statistical analysis.
58
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
Appendix 1. Etiology of hypertrophic pachymeningitis
Primary or idiopathic hypertrophic pachymeningitis Secondary causes Infectious diseases Mycobacterium tuberculosis Fungal infections Lyme disease Syphilis Eosinophilic meningitis due to parasitic diseases Systemic autoimmune/vasculitic disorders Sarcoidosis Wegener's granulomatosis Rheumatoid arthritis Behcet's disease Sjogren syndrome Temporal arteritis Mixed connective tissue disorder Orbital pseudotumor Vogt–Koyanagi–Harada disease IgG4 related inflammatory meningeal diseases Inflammatory bowel disease Others Dural carcinomatosis Meningioma Metastatic disease in adjacent skull bone Meningeal lymphoma POEMS (Polyneuropathy, Organomegaly, Endocrinopathy, Monoclonal plasma proliferative disorder, skin changes) Medications: Non-steroidal anti-inflammatory agents
Appendix 2a. Clinical scores of patients with hypertrophic pachymeningitis*
Headache 0 = no headache 1 = mild headache 2 = moderate 3 = severe 4 = very severe, disabled, confined to bed Encephalopathy 0 = absence of either seizure/encephalopathy 1 = presence of either encephalopathy/seizures Visual symptoms 0 = no visual symptoms 1 = present, not interfering with daily functioning 2 = present and interfering with daily functioning Ataxia 0 = no ataxia 1 = mild, able to move independently 2 = moderate, need 1 person's support 3 = severe, confined to bed/chair Hemiparesis/Quadriparesis 0 = no hemiparesis 1 = mild, able to move independently 2 = moderate, need 1 person's support 3 = severe, confined to bed/chair Other cranial nerve dysfunction** 0 = no symptoms 1 = present, not interfering with daily functioning 2 = present and interfering with daily functioning *Each symptom reported to be associated with pachymeningitis were scored separately and then the sum of the score was taken as the “composite score”. **For each cranial nerve dysfunction, the maximum score is 2 and the minimum score is 0 (except the optic and oculomotor nerves and the lower cranial nerves i.e. 9th, 10th, 11th & 12th nerves which are given scores taking them as a group) and thus the maximum total score for cranial nerve dysfunction is 12 and the maximum total clinical score is 23.
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
Appendix 2b. Clinical scores of patients with hypertrophic pachymeningitis
59
60
G.K. Dash et al. / Journal of the Neurological Sciences 350 (2015) 51–60
Appendix 3. The variability of clinical scores between clinician and patients
Variable
Kappa*
Level of agreement
Headache Encephalopathy Visual score Ataxia Hemiparesis Bulbar palsy Other cranial nerve score
0.529 1.0 0.836 1.0 1.0 1.0 0.906
Moderate Almost perfect Almost perfect Almost perfect Almost perfect Almost perfect Almost perfect
*Reference: Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33 [1]: 159–174.
References [1] Mamelak AN, Kelly WM, Davis RL, et al. Idiopathic hypertrophic cranial pachymeningitis. Report of 3 cases. J Neurosurg 1993;79:270–6. [2] Masson C, Hénin D, Hauw JJ, et al. Cranial pachymeningitis of unknown cause. A study of seven cases. Neurology 1993;43:1329–34. [3] Naffziger HC, Stern WE. Chronic pachymeningitis: report of a case and review of the literature. Arch Neuro Psychiatry 1949;62:383–411. [4] Parker Stephen W, Sobel Raymond A. Case 12—1988. N Engl J Med 1988;318:760–8. [5] Schwartz William J, Niles John L. Case 9—1999. N Engl J Med 1999;340:945–54. [6] Carruthers R, Carruthers M, Della-Torre E. IgG4-related disease and other causes of inflammatory meningeal disease. Semin Neurol 2014;34:395–404. [7] Lu LX, Della-Torre E, Stone JH, et al. IgG4-related hypertrophic pachymeningitis: clinical features, diagnostic criteria, and treatment. JAMA Neurol 2014;71:785–93.8. [8] Bonita R, Beaglehole R. Recovery of motor function after stroke. Stroke 1988;19: 1497–500. [9] Swieten JC, Koudstaal PJ, Visser MC, et al. Interobserver agreement for the assessment of handicap in stroke patients. Stroke 1988;19:604–7. [10] Stewart WF, Lipton RB, Kolodner K. Migraine disability assessment (MIDAS) score: relation to headache frequency, pain intensity, and headache symptoms. Headache 2003;43:258–65.
[11] Schmahmann JD, Gardner R, MacMore J, et al. Development of a brief ataxia rating scale (BARS) based on a modified form of the ICARS. Mov Disord 2009;24:1820–8. [12] Kupersmith MJ, Martin V, Heller G, et al. Idiopathic hypertrophic pachymeningitis. Neurology 2004;62:686–94. [13] Hatano N, Behari S, Nagatani T, et al. Idiopathic hypertrophic cranial pachymeningitis: clinicoradiological spectrum and therapeutic options. Neurosurgery 1999;45:1336–44. [14] Thomas B, Thamburaj K, Kesavadas C. ‘Eiffel-by-Night’: a new MR sign demonstrating reactivation in idiopathic hypertrophic pachymeningitis. Neuroradiol J 2007;20: 194–5. [15] Riku S, Kato S. Idiopathic hypertrophic pachymeningitis. Neuropathology 2003;23: 335–44. [16] Sylaja PN, Cherian PJ, Das CK, et al. Idiopathic hypertrophic cranial pachymeningitis. Neurol India 2002;50:53–9. [17] Della-Torre Galli L, Franciotta D, et al. Diagnostic value of IgG4 Indices in IgG4related hypertrophic pachymeningitis. J Neuroimmunol 2014:266:82–6. [18] Hamilton S, Smith C, Lessell S. Idiopathic hypertrophic cranial pachymeningitis. J Clin Neuroophthalmol 1993;13:127–34.
